© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    5. Llama3.1-8B (Base)
    7. Residual Stream
    9. 17-LLAMASCOPE-RES-131K
    11. 73669
    Prev
    Next
    INDEX
    Explanations

    phrases that suggest routines or actions involving "going for" something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Äįek
    -0.07
    orny
    -0.06
    çĽ
    -0.06
    enter
    -0.06
    elder
    -0.06
    ôi
    -0.06
    linkplain
    -0.06
    onse
    -0.06
    ovah
    -0.06
    et
    -0.06
    POSITIVE LOGITS
    Ware
    0.07
    .addListener
    0.06
    spin
    0.06
    illac
    0.06
     Walker
    0.06
    walker
    0.06
    PrototypeOf
    0.06
    olon
    0.06
     Kam
    0.06
    iddi
    0.06
    Activations Density 0.006%

    No Known Activations